|
Title:
|
FOCUSED TIME-DELAY NEURAL NETWORK MODELING TOWARDS TYPING STREAM PREDICTION |
|
Author(s):
|
Jun Li , Karim Ouazzane , Hassan Kazemian , Yanguo Jing , Richard Boyd |
|
ISBN:
|
978-972-8924-87-4 |
|
Editors:
|
António Palma dos Reis |
|
Year:
|
2009 |
|
Edition:
|
Single |
|
Keywords:
|
Focused Time-Delay Neural Network, Unary Coding, First Rank Conversion Values, Hitting Rate, First Hitting Rate, FT
Hitting Rate. |
|
Type:
|
Short Paper |
|
First Page:
|
189 |
|
Last Page:
|
194 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Users typing stream contains all the information of users interaction with computer by using QWERTY keyboard,
which may include user's vocabulary, typing habit and typing performance. This paper suggests a novel Focused Time-
Delay Neural Network model with extendable input and hidden layer neurons to analyze plain text and users historical
typing data. The experimental results indicate that about 50% First Three (FT) Hitting Rate is achieved, which reflect a
high FTDNN model performance. This can be applied to both, typing prediction and correction. |
|
|
|
|
|
|
